Prediction of defects in casting surface using image processing techniques.

Autor:	Balasubramani, S., Gokul, S., Ramakrishnan, T., Nitin, P., Prithivraj, K., Arjun, R. Naga
Předmět:	IMAGE recognition (Computer vision) METALLIC surfaces METAL defects SUPPORT vector machines K-means clustering
Zdroj:	AIP Conference Proceedings; 2024, Vol. 3221 Issue 1, p1-4, 4p
Abstrakt:	The method of flaw detection in hard reflective metal surfaces is made easier by developments in digital image processing techniques. Identification of metal surface defects is necessary for pattern recognition problems. One of the most often used classifiers for image classification is the Multi Class Support Vector Machine (Multi Class SVM) approach, which is used in this process. The obtained image undergoes preprocessing to eliminate any noise. When segmenting photos according to predetermined criteria such as size, texture, contrast, etc., K-means clustering is utilized. The suggested approach performs better when it comes to identifying the metal surface fault. [ABSTRACT FROM AUTHOR]
Databáze:	Complementary Index
Externí odkaz:	Zobrazit plný text záznamu